U.S. patent number 7,447,968 [Application Number 10/421,818] was granted by the patent office on 2008-11-04 for apparatus and method for supporting automatic repeat request in a high-speed wireless packet data communication system.
This patent grant is currently assigned to Samsung Electronics, Co., Ltd.. Invention is credited to Sang-Min Bae, Young-Kwon Cho, Sang-Hyuck Ha, Jin-Woo Heo, Min-Goo Kim.
United States Patent |
7,447,968 |
Ha , et al. |
November 4, 2008 |
Apparatus and method for supporting automatic repeat request in a
high-speed wireless packet data communication system
Abstract
Disclosed is an apparatus and method for supporting automatic
repeat request (ARQ) in a high-speed wireless packet data
communication system. A mobile station receives control information
including a subpacket identifier (SP_ID) representing the sequence
of a subpacket, the size of an encoder packet (EP_SIZE), and a
sequence identifier (AI_SN) that is toggled each time a new encoder
packet is transmitted, while receiving one of a plurality of
subpackets generated by segmenting a bit stream acquired by
encoding a transmission encoder packet. The mobile station
determines whether it will perform a decoding process caused by
initial transmission or a decoding process caused by retransmission
on the received subpacket, based on the control information.
Inventors: |
Ha; Sang-Hyuck (Suwon-shi,
KR), Kim; Min-Goo (Suwon-shi, KR), Heo;
Jin-Woo (Songnam-shi, KR), Cho; Young-Kwon
(Suwon-shi, KR), Bae; Sang-Min (Kyonggi-do,
KR) |
Assignee: |
Samsung Electronics, Co., Ltd.
(Suwon-Si, KR)
|
Family
ID: |
28793683 |
Appl.
No.: |
10/421,818 |
Filed: |
April 24, 2003 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20030202500 A1 |
Oct 30, 2003 |
|
Foreign Application Priority Data
|
|
|
|
|
Apr 24, 2002 [KR] |
|
|
10-2002-0022392 |
Feb 10, 2003 [KR] |
|
|
10-2003-0008263 |
|
Current U.S.
Class: |
714/748; 370/335;
370/342; 714/749; 714/750; 714/751 |
Current CPC
Class: |
H04L
1/0003 (20130101); H04L 1/0071 (20130101); H04L
1/1692 (20130101); H04L 1/1809 (20130101); H04L
1/1819 (20130101) |
Current International
Class: |
G08C
25/02 (20060101); H04L 1/18 (20060101) |
Field of
Search: |
;714/748-751 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
Lucent Technologies: "TSG-RAM#17(00)1382: Asynchronous and Adaptive
IR for HSDPA" Nov. 25, 2000; pp. 1-5. cited by other .
D.W. Krause, "PARQ-A System Supporting Information Networking Over
HF Radio Links"; Radio Systems and Techniques, Seventh
International Conference (Conf. Publ. No. 441), Jul. 7-10, 1997 pp.
408-412. cited by other .
Krishna R. Narayanan et al., "Physical Layer Design for Packet Data
Over IS-136" Vehicular Technology Conference, 1997 IEEE 47.sup.th,
vol. 2, May 4-7, 1997, pp. 102-1033, vol. 2. cited by other .
Qian Zhang et al., "Performance of a Type-II Hybrid ARQ Protocol in
Slotted DS-SSMA Packet Radio Systems", IEEE Transactions on
Communications, vol. 47, No. 2, Feb. 1999, pp. 281-290. cited by
other .
Y. Yuan-Wu, Performance of 64 kbit/s Data Transmission with the ARQ
Protocol of the DECT Standard on Fading Channels, Vehicular
Technology Conference, 2000, IEEE VTS-Fall VTC 2000, 52.sup.nd,
vol. 6, Sep. 24-28, 2000, pp. 2561-2567. cited by other .
Esa Malkamaki et al., "Performance of Hybrid ARQ Techniques for
WCDMA High Data Rates", Vehicular Technology Conference, 2001, IEEE
VTS 53.sup.rd, vol. 4, May 6-9, 2001, pp. 2720-2724. cited by
other.
|
Primary Examiner: Lamarre; Guy J
Attorney, Agent or Firm: Roylance, Abrams, Berdo &
Goodman, LLP
Claims
What is claimed is:
1. An automatic repeat request, ARQ, method for generating an
encoder packet by receiving control information including a
subpacket identifier, SP_ID, representing the sequence of a
subpacket, a size of an encoder packet EP_SIZE, and a sequence
identifier, AI_SN, that is toggled each time a new encoder packet
is transmitted, while receiving one of a plurality of subpackets
generated by segmenting a bit steam acquired by encoding a
transmission encoder packet, in a wireless packet data
communication system, the method comprising: generating the encoder
packet by performing a decoding process caused by initial
transmission on the received subpacket, if the AI_SN and the
EP_SIZE are both not identical to previously received AI_SN and
EP_SIZE; generating the encoder packet by performing the decoding
process caused by retransmission on the received subpacket, if the
AI_SN and the EP_SIZE are both identical to previously received
AI_SN and EP_SIZE and ACK is not transmitted in response to a
previously received subpacket.
2. The method of claim 1, further comprising: determining whether
the SP_ID is set to a value representing initial transmission, if
the AI_SN and the EP_SIZE are both identical to previously received
AI_SN and EP_SIZE and an acknowledge signal, ACK, is transmitted in
response to a previously received subpacket; and generating the
encoder packet by performing the decoding process caused by
retransmission on the received subpacket, if the SP_ID is not set
to the value representing initial transmission.
3. The method of claim 2, further comprising: generating the
encoder packet by performing the decoding process caused by initial
transmission on the received subpacket, if the SP_ID is set to the
value representing initial transmission.
4. The method of claim 1, wherein the step of performing the
decoding process caused by initial transmission comprises: setting
a reception number for the encoder packet to `0`, and clearing a
buffer; storing the received subpacket in the buffer, and
generating the encoder packet by decoding the stored subpacket; and
increasing the reception number by 1 if no error has occurred in
the encoder packet.
5. The method of claim 4, wherein the step of performing the
decoding process caused by retransmission comprises: combining the
received subpacket with a subpacket previously stored in the
buffer, and storing again the combined subpacket in the buffer;
generating the encoder packet by decoding the combined stored
subpacket; increasing the reception number by 1 if no error has
occurred in the encoder packet, and then determining whether the
reception number has arrived at a predetermined maximum reception
number; and clearing the buffer if the reception number has arrived
at the maximum reception number.
6. The method of claim 1, further comprising: determining whether
an error has occurred in the encoder packet; transmitting ACK if no
error has occurred; and transmitting a negative acknowledge signal,
NAK, if an error has occurred.
7. An automatic repeat request, ARQ, method for generating an
encoder packet by receiving control information including a
subpacket identifier, SP_ID, representing the sequence of a
subpacket, a size of an encoder packet EP_SIZE, and a sequence
identifier, AI_SN, that is toggled each time a new encoder packet
is transmitted, while receiving one of a plurality of subpackets
generated by segmenting a bit stream acquired by encoding a encoder
packet, in a wireless packet data communication system, the method
comprising: determining whether the SP_ID is set to a value
representing initial transmission; generating the encoder packet by
performing a decoding process caused by initial transmission on the
received subpacket, if the SP_ID is set to the value representing
initial transmission and the AI_SN and the EP_SIZE are both not
identical to previously received AI_SN and EP_SIZE; and generating
the encoder packet by performing a decoding process caused by
retransmission on the received subpacket, if the SP_ID is not set
to the value representing initial transmission and the AI_SN and
the EP_SIZE are both identical to previously received AI_SN and
EP_SIZE.
8. The method of claim 7, wherein the step of performing the
decoding process caused by initial transmission comprises: setting
a reception number for the encoder packet to `0`, and clearing a
buffer; storing the received subpacket in the buffer, and
generating the encoder packet by decoding the stored subpacket; and
increasing the reception number by 1 if no error has occurred in
the encoder packet.
9. The method of claim 8, wherein the step of performing the
decoding process caused by retransmission comprises: combining the
received subpacket with a subpacket previously stored in the
buffer, and storing again the combined subpacket in the buffer;
generating the encoder packet by decoding the combined stored
subpacket; increasing the reception number by 1 if no error has
occurred in the encoder packet, and then determining whether the
reception number has arrived at a predetermined maximum reception
number; and clearing the buffer if the reception number has arrived
at the maximum reception number.
10. The method of claim 7, further comprising the steps of:
determining whether an error has occurred in the encoder packet;
transmitting ACK if no error has occurred; and transmitting a
negative acknowledge signal, NAK, if an error has occurred.
11. An apparatus for receiving an encoder packet in a wireless
packet data communication system, comprising: a data channel
receiver for receiving, over a data channel, one of a plurality of
subpackets generated by segmenting a bit stream acquired by
encoding a transmission encoder packet, and generating an encoder
packet by performing a decoding process caused by initial
transmission or a decoding process caused by retransmission on the
received subpacket; a control channel receiver for receiving
control information including a subpacket identifier, SP_ID,
representing the sequence of the subpacket, a size of the encoder
packet, EP_SIZE, and a sequence identifier, AI_SN, that is toggled
each time a new encoder packet is transmitted; an acknowledge
channel, ACKCH, transmitter for transmitting an acknowledge signal,
ACK, if no error has occurred in the encoder packet, and
transmitting a negative acknowledge signal, NAK, if an error has
occurred in the encoder packet; and a hybrid automatic repeat
request, H-ARQ, controller for determining whether it will perform
the decoding process caused by initial transmission or the decoding
process caused by retransmission on the received subpacket, based
on the control information, such that the H-ARQ controller performs
the following operations: determining to perform the decoding
process caused by initial transmission on the received subpacket,
if the AI_SN and the EP_SIZE are both not identical to previously
received AI_SN and EP_SIZE; and determining to perform the decoding
process caused by retransmission on the received subpacket, if the
AI_SN and the EP_SIZE are both identical to previously received
AI_SN and EP_SIZE and ACK is not transmitted in response to a
previously received subpacket.
12. The apparatus of claim 11, wherein the H-ARQ controller
determines whether the SP_ID is set to a value representing initial
transmission, if the AI_SN and the EP_SIZE are both identical to
previously received AI_SN and EP_SIZE and ACK is transmitted in
response to a previously received subpacket, and determines to
perform the decoding process caused by retransmission on the
received subpacket, if the SP_ID is not set to the value
representing initial transmission.
13. The apparatus of claim 12, wherein the H-ARQ controller
determines to perform the decoding process caused by initial
transmission on the received subpacket, if the SP_ID is set to the
value representing initial transmission.
14. The apparatus of claim 11, wherein if the H-ARQ controller
determines to perform the decoding process caused by initial
transmission, the data channel receiver comprises: setting a
reception number for the encoder packet to `0`, and clearing a
buffer; storing the received subpacket in the buffer, and
generating the encoder packet by decoding the stored subpacket; and
increasing the reception number by 1 if no error has occurred in
the encoder packet.
15. The apparatus of claim 14, wherein if the H-ARQ controller
determines to perform the decoding process caused by
retransmission, the data channel receiver comprises: combining the
received subpacket with a subpacket previously stored in the
buffer, and storing again the combined subpacket in the buffer;
generating the encoder packet by decoding the combined stored
subpacket; if no error has occurred in the encoder packet,
increasing the reception number by 1 and then determining whether
the reception number has arrived at a predetermined maximum
reception number; and clearing the buffer if the reception number
has arrived at the maximum reception number.
16. An apparatus receiving an encoder packet in a wireless packet
data communication system, comprising: a data channel receiver for
receiving, over a data channel, one of a plurality of subpackets
generated by segmenting a bit stream acquired by encoding a
transmission encoder packet, and generating an encoder packet by
performing a decoding process caused by initial transmission or a
decoding process caused by retransmission on the received
subpacket; a control channel receiver for receiving control
information including a subpacket identifier, SP_ID, representing
the sequence of the subpacket, a size of the encoder packet,
EP_SIZE, and a sequence identifier, AI_SN, that is toggled each
time a new encoder packet is transmitted; an acknowledge channel,
ACKCH, transmitter for transmitting an acknowledge signal, ACK, if
no error has occurred in the encoder packet, and transmitting a
negative acknowledge signal, NAK, if an error has occurred in the
encoder packet; and a hybrid automatic repeat request, H-ARQ,
controller for determining whether it will perform the decoding
process caused by initial transmission or the decoding process
caused by retransmission on the received subpacket, based on the
control information, the H-ARQ controller performing the following
operations: determining whether the SP_ID is set to a value
representing initial transmission; determining to perform the
decoding process caused by initial transmission on the received
subpacket if the SP_ID is set to the value representing initial
transmission and the AI_SN and the EP_SIZE are both not identical
to previously received AI_SN and EP_SIZE; and determining to
perform the decoding process caused by retransmission on the
received subpacket if the SP_ID is not set to a value representing
initial transmission and the AI_SN and the EP_SIZE are both
identical to previously received AI_SN and EP_SIZE.
17. The apparatus of claim 16, wherein if the H-ARQ controller
determines to perform the decoding process caused by initial
transmission, the data channel receiver comprises: setting a
reception number for the encoder packet to `0`, and clearing a
buffer; storing the received subpacket in the buffer, and
generating the encoder packet by decoding the stored subpacket; and
increasing the reception number by 1 if no error has occurred in
the encoder packet.
18. The apparatus of claim 17, wherein if the H-ARQ controller
determines to perform the decoding process caused by
retransmission, the data channel receiver comprises: combining the
received subpacket with a subpacket previously stored in the
buffer, and storing again the combined subpacket in the buffer;
generating the encoder packet by decoding the combined stored
subpacket; if no error has occurred in the encoder packet,
increasing the reception number by 1 and then determining whether
the reception number has arrived at a predetermined maximum
reception number; and clearing the buffer if the reception number
has arrived at the maximum reception number.
Description
PRIORITY
This application claims priority under 35 U.S.C. .sctn. 119 to an
application entitled "Apparatus and Method for Supporting Automatic
Repeat Request in a High-Speed Wireless Packet Data Communication
System" filed in the Korean Intellectual Property Office on Apr.
24, 2002 and assigned Serial No. 2002-22392, and an application
entitled "Apparatus and Method for Supporting Automatic Repeat
Request in a High-Speed Wireless Packet Data Communication System"
filed in the Korean Intellectual Property Office on Feb. 10, 2003
and assigned Serial No. 2003-8263, the contents of both of which
are incorporated herein by reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a high-speed wireless
packet data communication system, and in particular, to an
apparatus and method for supporting automatic repeat request.
2. Description of the Related Art
Wireless communication systems have been developed to enable users
to continue smooth communication while travelling. The development
of a wireless communication technology has given impetus to
research in technology for transmitting large quantities of data to
users.
Wideband Code Division Multiple Access (W-CDMA) wireless
communication system a typical CDMA2000 1.times. system, were
designed to support only voice service and relatively low-speed
packet data service. However, with the development of communication
technology and at the request of users, research has been carried
out in wireless communication systems supporting a high-speed
packet data service. Among others, one example is an IS-2000
1.times.EV-DV (Evolution in Data and Voice) system, which is the
new CDMA standard developed by 3GPP2 (3rd Generation Partnership
Project 2). This is considered as the preferred system for
supporting not only a voice service, but also high-speed packet
data service. In order to realize a system capable of supporting
both high-speed packet data service as well as voice service, it is
necessary to design both a base station (BS) and mobile station
(MS) capable of handling high-speed wireless packet data.
In a wireless channel environment characterized in that variation
in channel conditions are considerable and traffic channels for
different types of services coexist, automatic repeat request
(hereinafter referred to as "ARQ") is used to increase transmission
efficiency, or transmission throughput, for high-speed data
transmission. According to ARQ protocols, a receiver requests
retransmission when an error has occurred in received data, and a
transmitter retransmits the data in response to the request so that
the receiver can obtain reception quality over a predetermined
level.
In a typical wireless communication system, ARQ could not be
considered for a voice call service, since the voice call service
is requested real-time processing. Thus, ARQ was restrictively
supported in an upper layer for a circuit-based data service only.
As a proposal for increasing transmission throughput, a proposal
has been made for supporting ARQ in a physical layer rather than in
the upper layer, and this proposal is discussed as a part of the
standard for a high-speed wireless packet data communication system
(e.g., 1.times.EV-DV system) that is currently being developed.
In a currently discussed system, a receiver corrects an error
existing in received data within the limit of its error correction
capability, but sends a retransmission request message to a
transmitter when it cannot correct the error: this technology is
called hybrid ARQ (hereinafter referred to as "H-ARQ"). Since
commercialization of a high-speed data transmission service will
soon be available, analysis and research has been conducted on a
technology for efficiently applying, in an actual system, new H-ARQ
where a variable coding rate error correction code is used rather
than existing hybrid H-ARQ where a fixed coding rate error
correction code is used. Furthermore, at least one study has been
made of a method for applying, to the system, high level modulation
e.g., 8-ary phase shift keying (8PSK), 16-ary quadrature amplitude
modulation (16QAM), binary phase shift keying (BPSK) and quadrature
phase shift keying (QPSK), for a high-speed transmission channel
structure.
The new CDMA standard, the IS-2000 1.times.EV-DV system adopts a
coding scheme using quasi-complementary turbo codes (QCTC). The
QCTC supports the variable coding rate for H-ARQ on high-speed
data, and secures improvement of soft-combining performance by
H-ARQ. In the 1.times.EV-DVsystem, transmission and reception of
packet data is performed by H-ARQ or fast H-ARQ of a physical
layer.
When implementing H-ARQ, many factors should be considered in terms
of system complexity, such as buffer size and signaling load of a
transmitter and a receiver, in addition to the transmission
throughput. Accordingly, there have been demands for a control
algorithm for more efficiently managing H-ARQ for improving system
efficiency and service quality in a high-speed wireless packet data
communication system.
SUMMARY OF THE INVENTION
It is, therefore, an object of the present invention to provide an
apparatus and method for controlling an H-ARQ protocol in a
high-speed wireless packet data communication system.
It is another object of the present invention to provide an
apparatus and method for transmitting a response to a packet
received according to an H-ARQ protocol in a high-speed wireless
packet data communication system.
It is a further object of the present invention to provide an
apparatus and method for separately handling an initially
transmitted packet and a retransmitted packet, both received
according to an H-ARQ protocol, in a high-speed wireless packet
data communication system.
In accordance with one aspect of the present invention, there is
provided an automatic repeat request (ARQ) method for generating an
encoder packet by receiving control information including a
subpacket identifier (SP_ID) representing sequence of a subpacket,
a size (EP_SIZE) of an encoder packet, and a sequence identifier
(AI_SN) that is toggled each time a new encoder packet is
transmitted, while receiving one of a plurality of subpackets
generated by segmenting a bit stream acquired by encoding a
transmission encoder packet, in a wireless packet data
communication system. The ARQ method comprises generating the
encoder packet by performing a decoding process caused by initial
transmission on the received subpacket, if the AI_SN and the
EP_SIZE are both not identical to previously received AI_SN and
EP_SIZE; determining whether the SP_ID is set to a value
representing initial transmission, if the AI_SN and the EP_SIZE are
both identical to previously received AI_SN and EP_SIZE and an
acknowledge signal (ACK) is transmitted in response to a previously
received subpacket; generating the encoder packet by performing a
decoding process caused by initial transmission or retransmission
on the received subpacket according to the SP_ID; and generating
the encoder packet by performing the decoding process caused by
retransmission on the received subpacket, if the AI_SN and the
EP_SIZE are both identical to previously received AI_SN and EP_SIZE
and ACK is not transmitted in response to a previously received
subpacket.
BRIEF DESCRIPTION OF THE DRAWINGS
The above and other objects, features and advantages of the present
invention will become more apparent from the following detailed
description when taken in conjunction with the accompanying
drawings in which:
FIG. 1 illustrates a schematic structure of an F-PDCH transmitter
for a packet data service;
FIG. 2 illustrates a schematic structure of an F-PDCH receiver
corresponding to the F-PDCH transmitter of FIG. 1;
FIG. 3 is a block diagram illustrating a relationship between an
upper layer and a physical layer for typical H-ARQ processing;
FIG. 4 is a block diagram illustrating an improved relationship
between an upper layer and a physical layer for H-ARQ
processing;
FIG. 5 is a block diagram illustrating a hierarchical structure of
a mobile station according to a preferred embodiment of the present
invention;
FIG. 6 is a flowchart illustrating an operation of detecting an
H-ARQ response by a base station employing a typical tri-state
detection scheme;
FIG. 7 is a flowchart illustrating an H-ARQ operation by a base
station using MAC_ID, ARQ_ID, SP_ID and EP_SIZE according to an
embodiment of the present invention;
FIG. 8 is a flowchart illustrating an H-ARQ operation by a mobile
station using MAC_ID, ARQ_ID, SP_ID and EP_SIZE according to an
embodiment of the present invention;
FIG. 9 is a flowchart illustrating an H-ARQ operation by a base
station using MAC_ID, ARQ_ID, SP_ID, EP_SIZE and AI_SN according to
an embodiment of the present invention;
FIG. 10 is a flowchart illustrating an H-ARQ operation by a mobile
station using MAC_ID, ARQ_ID, SP_ID, EP_SIZE and AI_SN according to
an embodiment of the present invention;
FIG. 11 is a flowchart illustrating an H-ARQ operation by a base
station according to another embodiment of the present
invention;
FIG. 12 is a flowchart illustrating an H-ARQ operation by a mobile
station according to another embodiment of the present
invention;
FIGS. 13A to 13D are signal flow diagrams for normal situations
according to an H-ARQ protocol;
FIGS. 14A to 14E are signal flow diagrams for abnormal situations
according to an H-ARQ protocol; and
FIG. 15 is a flowchart illustrating an H-ARQ operation by a mobile
station according to an additional embodiment of the present
invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Several preferred embodiments of the present invention will now be
described in detail with reference to the annexed drawings. In the
drawings, the same or similar elements are denoted by the same
reference numerals even though they are depicted in different
drawings. In the following description, a detailed description of
known functions and configurations incorporated herein has been
omitted for conciseness.
A description will now be made of major channels necessary for a
high-speed packet data service according to the present invention.
The term "forward channel" as used herein refers to a channel
configured in a direction of a base station to a mobile station,
and the term "reverse channel" refers to a channel configured in a
direction of a mobile station to a base station. In most cases,
large quantities of packet data are transmitted from a base station
to a mobile station, so the present invention will be described
with reference to forward data transmission. Therefore, in the
following description, it should be understood that a transmitter
refers to a base station and a receiver refers to a mobile station.
Of course, the present invention can be applied to all kinds of
systems that transmit and receive data wirelessly, without being
limited to data transmission from a base station to a mobile
station.
Forward channels for a packet data service are classified into a
common channel, a control channel, and a traffic channel. In the
following description, a prefix "F-" of a particular channel means
a forward link configured in a direction of a base station to a
mobile station, while a prefix "R-" of a particular channel means a
reverse link configured in a direction of a mobile station to a
base station.
The common channel represents a pilot channel (PICH), and provides
reference amplitude and phase variation for synchronous
demodulation at a mobile station. The traffic channel includes a
packet data channel (PDCH) for actually transmitting packet data,
and the control channel includes a forward packet data control
channel (PDCCH) for transmitting control information related to
reception of the packet data channel.
The control information includes the following types of
information: medium access control identifier (MAC_ID) representing
a destination mobile station to which a transmission packet is
forwarded; a subpacket identifier (SP_ID) representing a
retransmission number of a transmission packet; an ARQ
identification sequence number (AI_SN) which is toggled at each
transmission of a new packet, in order to indicate whether a
transmission packet is an even-numbered packet or an odd-numbered
packet; an ARQ identifier (ARQ_ID) representing a channel over
which a transmission is transmitted, among ARQ channels transmitted
in parallel; an encoder packet size (EP_SIZE) representing a size
of a transmission packet; a Walsh space indicator representing
Walsh codes to be used for PDCH, and a code division multiplexing
(CDM) channel identifier.
A typical scheme for transmitting and receiving a forward packet
data channel (F-PDCH) for a packet data service will be described
herein below.
FIG. 1 illustrates a schematic structure of an F-PDCH transmitter
for a packet data service. Herein, the F-PDCH transmitter refers to
a base station.
Referring to FIG. 1, an input sequence of F-PDCH is applied to a
cyclic redundancy code (CRC) adder 10, and the CRC adder 10 adds a
16-bit CRC to the input sequence. The CRC-added input sequence is
encoded by a channel encoder 11. Herein, the input sequence is
called an "encoder packet" (EP), since it is encoded by the channel
encoder 11. The channel encoder 11 has a predetermined coding rate
R for encoding the encoder packet. If the coding rate R is k/n (n
and k are relatively prime), the channel encoder 11 outputs n bits
for a k-bit input encoder packet. For example, the coding rate can
be 1/2 or 3/4.
The channel encoder 11 employs turbo coding that is estimated to be
the most proper coding for reliable high-speed transmission of
multimedia data in a next generation mobile communication system.
The encoder packet encoded by turbo coding is divided into a
systematic part and a parity part. The systematic part means
information itself to be transmitted, and the parity part means
error correction information added at the transmitter during
decoding, to correct an error which occurred during
transmission.
For example, if the coding rate R is a symmetric coding rate of
1/2, the channel encoder 11 generates a 2-bit coded output in
response to a 1-bit input sequence, and the coded bits include a
1-bit systematic part (information bit) and a 1-bit parity part
(parity bit). As another example, if the coding rate R is an
asymmetric coding rate of 3/4, the channel encoder 11 generates a
4-bit coded output in response to a 3-bit input sequence, and the
coded bits include a 3-bit systematic part and a 1-bit parity
part.
In particular, when H-ARQ is used, the coded bit stream
corresponding to each input encoder packet is divided into a
predetermined number of subpackets each having different increment
redundancy (IR) patterns by QCTC symbol selection, and the channel
encoder 11 sequentially generates the subpackets one by one, each
time a retransmission request for the encoder packet is received.
In practice, the transmitter transmits different subpackets at each
retransmission. Since all of these subpackets correspond to the
same encoder packet, it will be assumed herein that transmission of
subpackets corresponding to the same encoder packet is equivalent
to retransmission of an encoder packet. The subpackets are
identified by SP_ID, and are transmitted from a base station over
the F-PDCCH. The SP_ID represents the number of retransmissions on
the same encoder packet.
A method of generating the subpackets or sequence in which the
subpackets are transmitted, is determined according to a type of
H-ARQ applied thereto. Therefore, a plurality of subpackets
corresponding to the same encoder packet can be either identical
to, or different from one another. For example, a first subpacket,
or an initially transmitted subpacket, is comprised of a part or
the whole of a systematic part in the coded bit stream, and
retransmitted subpackets may include only a parity part, or a part
or the whole of a systematic part according to the retransmission
number. In the following description, the first subpacket will be
referred to as "new subpacket" (or initial transmission subpacket),
and its succeeding subpackets will be referred to as "continue
subpackets" (retransmission subpackets).
The coded output of the channel encoder 11 undergoes rate matching
by a rate matcher 12. Generally, the rate matching is performed by
repetition and puncturing on the coded output, when a transport
channel is multiplexed or the number of the coded output bits is
not identical to the number of bits that can be transmitted over
the air.
The rate-matched output undergoes interleaving by an interleaver
13. The interleaving is performed so that a damaged part of the
transmission bits should be dispersed instead of being concentrated
upon a particular position, thereby preventing a possible burst
error which frequently occurs while the transmission bits pass
through a fading channel. The prevention of a burst error
contributes to minimization of a transmission loss and an increase
in a channel coding gain.
The interleaved output is mapped to modulation symbols according to
a particular modulation order selected from QPSK, 8PSK, 16QAM and
64QAM by an M-ary modulator 14. The modulation order is selected
according to a current condition of the wireless channel.
Though not illustrated in FIG. 1, the transmitter spreads the
modulation symbols with a pseudo-random noise (PN) code for
identification of a base station and a plurality of Walsh codes for
identification of transmission channels, so that a receiver can
identify the data transmission channels and the base station that
transmits the data.
FIG. 2 illustrates a schematic structure of an F-PDCH receiver
corresponding to the F-PDCH transmitter of FIG. 1. Herein, the
F-PDCH receiver refers to a mobile station.
Referring to FIG. 2, received data is despread with a PN code for
identification of the transmitting base station and a plurality of
Walsh codes for identification of transmission channels by a
demodulator 20. The demodulator 20 demodulates the received data by
a demodulation scheme corresponding to the modulation scheme used
in the modulator 14 of the transmitter. A deinterleaver 21
deinterleaves the demodulated output according to the
deinterleaving rule corresponding to the modulating rule used in
the interleaver 13 of the transmitter, thereby outputting
subpackets. Herein, the "subpacket" refers to each of a plurality
of transmission units that are retransmitted for the same encoder
packet, and is identified by the SP_ID transmitted from the base
station over F-PDCCH. A combiner 22 combines subpackets accumulated
in an ARQ buffer 26 for the same encoder packet with the subpackets
output from the deinterleaver 21 according to the type of H-ARQ
employed. If there is no subpacket accumulated for the same encoder
packet, the subpackets output from the interleaver 21 are output
without combining. An output of the combiner 22 is provide to a
channel decoder 23, and at the same time, stored in the ARQ buffer
26 so that it can be combined with a next received subpacket.
The channel decoder 23 decodes the output of the combiner 22 by a
predetermined decoding scheme, thereby restoring a desired encoder
packet. The decoding scheme is determined based on the coding
scheme performed in the channel encoder 11 of the transmitter.
A CRC checker 24 extracts a CRC from the encoder packet decoded by
the channel decoder 23, and determines whether the encoder packet
has an error based on the extracted CRC. As a result of the
determination, the CRC checker 24 transmits an acknowledge signal
(ACK) indicating correct receipt of the encoder packet or a
non-acknowledge signal (NAK) indicating retransmission request for
the encoder packet, to the base station over a reverse acknowledge
channel (R-ACKCH).
When ACK is transmitted, buffer initialization is performed to
delete subpackets stored in the ARQ buffer 26 for the corresponding
encoder packet. On the contrary, if NAK is transmitted,
corresponding subpackets are retained in the ARQ buffer 26.
Format information necessary for reception of the F-PDCH configured
in the stated-above manner can be acquired from control information
transmitted over F-PDCCH. The control information, as mentioned
above, includes MAC_ID, SP_ID, AI_SN, ARQ_ID and EP_SIZE. A mobile
station determines whether MAC_ID acquired by demodulating F-PDCCH
is identical to its own MAC_ID. If the acquired MAC_ID is identical
to its own MAC_ID, the mobile station activates an F-PDCH receiver
to receive F-PDCH data, and then transmits a decoding result for
the received data to the base station over the R-ACKCH.
The above-described transmitter and receiver correspond to physical
layers of a base station and a mobile station, respectively. In a
typical wireless communication system with a hierarchical
structure, H-ARQ is achieved in an upper layer.
FIG. 3 is a block diagram illustrating a relationship between an
upper layer and a physical layer for typical H-ARQ processing.
Referring to FIG. 3, a physical layer 32 decodes a signal
transmitted over a wireless channel, and delivers the decoded data
to a MAC layer 30, the upper layer, along with its CRC check result
in the form of MuxPDU (Multiplexed Protocol Data Unit). The MAC
layer 30 determines from the CRC check result whether the decoded
data from the physical layer 32 has an error. If no error has
occurred, the MAC layer 30 requests transmission of new data.
Further, the MAC layer 30 handles the data according to a radio
link protocol (RLP). In this manner, the data actually transmitted
from the transmitter is determined by the MAC layer 30.
In the case where H-ARQ processing is achieved by the MAC layer 30,
the data decoded by the physical layer 32 must be delivered to the
MAC layer 30, the upper layer, thus causing a decrease in a
processing speed for each retransmission data and an increase in a
load of the MAC layer 30 during high-speed data processing.
In addition, when H-ARQ processing is performed in the upper layer,
soft combining on the same data cannot be performed. This is
because although the physical layer can retain a soft value for
each received demodulated symbol, the symbols delivered to the
upper layer are converted into a hard value which is a binary value
with all 0s or all 1s. Therefore, even though coded symbols for the
same encoder packet are repeatedly received by retransmission, the
coded symbols cannot be soft-combined. The only available method
for symbol combining is to calculate the number of 0s or 1s for the
symbols having a binary value, compare the number of 0s with the
number of 1s, and select major symbols. This method is called
"majority voting". However, majority voting is also scarcely used
due to the required amount of calculations. On the contrary, if
H-ARQ processing is performed in the physical layer, coded symbols
for the same encoder packet can be soft-combined, securing
efficient resource utilization.
For these reasons, in accordance with an embodiment of the
invention, a structure for handling a part of an H-ARQ operation in
a physical layer is proposed as illustrated in FIG. 4.
FIG. 4 is a block diagram illustrating an improved relationship
between an upper layer and a physical layer for H-ARQ processing.
Referring to FIG. 4, for fast processing and response for H-ARQ, a
part of an H-ARQ operation that was conventionally performed in a
MAC layer 40 is performed in a physical layer 44 or a sublayer 42
intervening between the MAC layer 40 and the physical layer 44.
Herein, the sublayer 42 will be referred to as H-ARQ control
sublayer 42. The H-ARQ control sublayer 42 performs a part of the
operation that was conventionally performed in the MAC layer 30 of
FIG. 3. The H-ARQ control sublayer 42 is structurally included in
the physical layer 44, and functionally included in the MAC layer
40. The H-ARQ control sublayer 42 is structured so that it should
perform only the operation of determining whether to retransmit
data on behalf of the MAC layer 40, thereby reducing H-ARQ
processing time for the same data.
The improved structure of FIG. 4 increases H-ARQ processing speed,
compared with the conventional structure. The structure of FIG. 3
receives a NAK signal through one packet transmission, and round
trip delay of a minimum of about 200 msec occurs until a time point
where a retransmission packet is transmitted. However, in the
structure of FIG. 4, short round trip delay of about several
milliseconds occurs. In the case where F-PDCH uses adaptive
modulation and coding (AMC) that changes a modulation scheme and a
coding rate according to a channel condition, such short round trip
delay is particularly useful. In order to substantially manage
H-ARQ using the interlayer structure between the upper layer and
the physical layer, a retransmission protocol for a transmitter
according to a retransmission request (i.e., NAK transmitted from a
receiver) is required. For this, a CDMA2000 1.times. system
specified by 3GPP2 uses AAIR (Asynchronous and Adaptive Increment
Redundancy). According to AAIR, a base station asynchronously
performs packet transmission to a mobile station depending on
forward channel quality information reported to the mobile station,
and a modulation scheme, a coding rate and a slot length of the
transmission packet are adaptively determined according to the
channel quality. The packet, initial transmission of which failed,
is retransmitted, and a code symbol pattern (or subpacket)
transmitted at retransmission can be different from a code symbol
pattern transmitted at initial transmission. AAIR increases the
signal-to-noise ratio (SNR) of packet data according to an increase
in the number of retransmissions, and increases the coding gain in
proportion according to a decrease in a coding rate, thereby
improving packet data transmission/reception performance.
FIG. 5 is a block diagram illustrating a hierarchical structure of
a mobile station according to a preferred embodiment of the present
invention. In FIG. 5, an H-ARQ controller 50 corresponds to the
H-ARQ control sublayer 42, and the R-ACKCH transmitter 52, F-PDCCH
receiver 54 and F-PDCH receiver 56 correspond to the physical layer
44. The F-PDCH receiver 56 is identical in structure to the
receiver described in conjunction with FIG. 2.
Referring to FIG. 5, the R-ACKCH transmitter 52 generates an ACK
signal or a NAK signal to be transmitted over R-ACKCH, under the
control of the H-ARQ controller 50. The F-PDCCH receiver 54 decodes
control information received over F-PDCCH, and provides the decoded
result to the H-ARQ controller 50 and the F-PDCH receiver 56. The
F-PDCH receiver 56, connected to an interlayer buffer 58 for
received packets, receives each packet data received over F-PDCH,
and performs demodulation and decoding on the received packet data
under the control of the H-ARQ controller 50. The data successfully
decoded by the F-PDCH receiver 56 is stored in the interlayer
buffer 58 under the control of the H-ARQ controller 50. The
interlayer buffer 58, under the control of the H-ARQ controller 50,
delivers the stored data to an upper layer at periods determined
such that the upper layer does not have a large load.
The H-ARQ controller 50 determines whether it will perform
retransmission or initial transmission on packet data received by
the F-PDCH receiver 56, depending on control information MAC_ID,
ARQ_ID, SP_ID, EP_SIZE and AI_SN received by the F-PDCCH receiver
54, and controls the R-ACKCH transmitter 52 according to the
processing result thereby to generate an ACK or NAK signal.
A binary detection scheme for detecting ACK and NAK, or a tri-state
detection scheme for detecting ACK, NAK and Silence, can be used as
a detection scheme in which a mobile station informs a base station
of success in decoding an encoder packet over R-ACKCH. The term
"Silence" means actual no-response where R-ACKCH is gated off.
FIG. 6 illustrates a procedure for detecting ACK, NAK or Silence by
an R-ACKCH receiver of a base station in the case where the
tri-state detection scheme is adopted. Referring to FIG. 6, a base
station extracts an R-ACKCH signal by despreading a received signal
with a spreading code assigned to R-ACKCH (Step 60), and then
detects a power (or energy) level by separating the R-ACKCH signal
into an in-phase (I) channel signal and a quadrature-phase (Q)
channel signal and summing up squares of the separated channel
signals (Step 62). The detected power level is compared with a
predetermined threshold (Step 64). If the detected power level is
lower than the threshold, it means that R-ACKCH is gated off by a
mobile station, so the base station detects Silence (Step 66).
However, if the detected power level is higher than or equal to the
threshold, the base station decodes the R-ACKCH signal, determines
whether the decoding result corresponds to ACK or NAK, and then
transmits a new subpacket for the same encoder packet
(retransmission process) or transmits an initial transmission
subpacket for a new encoder packet (initial transmission process)
according to the determination result (Step 68).
As described above, a base station can repeatedly transmit the same
encoder packet according to a response from a mobile station. In
this case, the mobile station should determine whether a currently
received packet is a retransmitted packet or an initially
transmitted packet, and perform a process corresponding thereto. If
the mobile station determines whether the currently received packet
is a retransmitted packet depending only on whether it has
previously transmitted an ACK or NAK, the mobile station may
incorrectly interpret a response to R-ACKCH. As a result, the
mobile station may perform unnecessary processing or mistakenly
discard a received packet.
Therefore, in accordance with an embodiment of the present
invention, a mobile station determines whether it will perform
retransmission or initial transmission on a currently received
packet depending upon a previous response to R-ACKCH and/or control
information on F-PDCCH. Retransmission-related information included
in the control information on F-PDCCH includes; ARQ_ID indicating
an ARQ channel to which a transmission packet belongs; EP_SIZE
indicating a size of an encoder packet; SP_ID indicating sequence
of a subpacket; and an AI_SN flag indicating initial transmission
(NEW) or retransmission (CONTINUE) of a transmission packet.
With reference to the accompanying drawings, a description will now
be made as to how a base station and a mobile station perform H-ARQ
processing on one of plural ARQ channels based on
retransmission-related information on F-PDCCH. Each ARQ channel is
processed in the same way.
FIGS. 7 to 12 illustrate the operation of performing H-ARQ
processing in the case where a mobile station should necessarily
receive an initially transmitted subpacket (SP_ID=0). As mentioned
before, an initially transmitted subpacket includes a part or the
whole of a systematic part in a coded bit stream. For successful
decoding, the systematic part is relatively important as compared
with a parity part. In FIGS. 7 to 12, when a mobile station misses
F-PDCCH or misses an initially transmitted subpacket on F-PDCCH,
the base station retransmits the initially transmitted subpacket in
order to improve a decoding success rate of an encoder packet.
FIGS. 7 to 12 illustrate an operation of performing H-ARQ
processing depending upon elements selected from control
information transmitted from a base station. Specifically, FIGS. 7
and 8 illustrate an operation of performing H-ARQ processing
depending upon other elements except AI_SN.
FIGS. 7 and 8 are flowcharts illustrating H-ARQ operations by a
base station and a mobile station depending upon MAC_ID, ARQ_ID,
SP_ID and EP_SIZE according to an embodiment of the present
invention. Herein, SP_ID=0 indicates initial transmission (or first
transmission) on a first subpacket of a new encoder packet.
Referring to FIG. 7, a base station sets a transmission number
Tx_Num to 0 (Step 100), and determines whether the transmission
number Tx_Num is 0 in order to start a packet data service (Step
102).
If the transmission number Tx_Num is 0 ("Yes" path from decision
step 102), the base station transmits control information including
MAC_ID, ARQ_ID, SP_ID (herein SP_ID is set to 0) and EP_SIZE over
F-PDCCH while transmitting an initial transmission subpacket of a
new encoder packet to a mobile station over F-PDCH (Step 104), and
then waits for a response from the mobile station (Step 106).
Thereafter, the base station decodes R-ACKCH from the mobile
station (Step 108), and analyzes the response to the transmitted
subpacket from the mobile station (Step 110). If the response from
the mobile station is ACK ("Yes") path from decision step 110), the
base station returns to step 102 to transmit a new encoder packet.
Otherwise, if the response from the mobile station is NAK or
Silence, the base station increases the transmission number Tx_Num
by 1 ("No" path from decision step 110) in order to retransmit the
same encoder packet (Step 112), and then returns to step 102.
If it is determined in step 102 that the transmission number Tx_Num
is not 0 ("No" path from decision step 102), the base station
determines whether the last response received from the mobile
station over R-ACKCH is Silence (Step 114). If the last response
from the mobile station is Silence, i.e., if there is no previous
response ("No" path from decision step 114), the base station
transmits control information including previous SP_ID over F-PDCCH
while retransmitting a previously transmitted subpacket over F-PDCH
(Step 118). However, if the last response from the mobile station
is not Silence, i.e., if there is a previous response ("Yes" path
from decision step 114), the base station increases SP_ID of
control information transmitted over F-PDCCH by 1 while
transmitting a next subpacket of a previously transmitted encoder
packet, over F-PDCH (Step 116). The SP_ID is set to a value
indicating a retransmission number for the same encoder packet. For
example, SP_ID is set to `01` for first retransmission, and `10`
for second retransmission.
Thereafter, the base station waits for a response from the mobile
station for the subpacket transmitted in step 116 or 118 (Step
120). The base station decodes R-ACKCH from the mobile station
(Step 122), and analyzes a response from the mobile station for the
subpacket transmitted in step 116 or 118 (Step 124). If the
response from the mobile station is ACK ("Yes" path from decision
step 124), the base station resets the transmission number Tx_Num
to `0` in order to transmit a new encoder packet (Step 130), and
then returns to step 102. However, if a response from the mobile
station is NAK or Silence ("No" path from decision step 124), the
base station increases the transmission number Tx_Num by 1 (Step
126), and then proceeds to step 128.
In step 128, the transmission number Tx_Num is compared with a
previously set maximum transmission number MAX_TX_NUM. The maximum
transmission number MAX_TX_NUM prevents the base station from
infinitely retransmitting the same packet data. If the transmission
number Tx_Num is larger than or equal to the maximum transmission
number MAX_TX_NUM ("Yes" path from decision step 128), the base
station resets the transmission number Tx_Num to `0` in order to
transmit a new encoder packet (Step 130), and then returns to step
102. However, if the transmission number Tx_Num is smaller than the
maximum transmission number MAX_TX_NUM ("No" path from decision
step 128), the base station returns to step 102 without changing
the transmission number Tx_Num in order to retransmit the same
encoder packet.
Referring to FIG. 8, a mobile station initializes an ARQ buffer and
sets a reception number Rx_Num to 0 in order to receive a packet
data service (Step 200). Thereafter, the mobile station decodes
F-PDCCH from a base station (Step 202), and determines whether a
CRC error is detected from control information acquired by decoding
the F-PDCCH, thereby determining whether the mobile station made a
successful in decoding (Step 204). The control information acquired
by decoding the F-PDCCH includes MAC_ID, SP_ID and EP_SIZE. If the
decoding failed ("No" path from decision step 204), R-ACKCH is
gated off (Step 206). In this case, nothing is transmitted over the
R-ACKCH, and this is regarded as Silence by the base station.
If it is determined in step 204 that decoding of F-PDCCH was
successfully performed ("Yes" path from decision step 204), the
mobile station determines whether the MAC_ID acquired by the
decoding is identical to its own MAC_ID, thereby determining
whether there is any packet data transmitted thereto (Step 208). If
they are not identical to each other ("No" path from decision step
208), R-ACKCH is gated off (Step 206).
Otherwise, if the MAC_IDs are identical to each other in step 208
("Yes" path from decision step 208), the mobile station determines
whether SP_ID acquired by the decoding is set to `0` that indicates
initial transmission (Step 210). If the SP_ID is set to `0` ("Yes"
path from decision step 210), the mobile station sets the reception
number Rx_Num to `0` (Step 212), and then clears the ARQ buffer
(Step 214). Thereafter, the mobile station stores a subpacket
received over F-PDCH, in the ARQ buffer, and acquires an encoder
packet by decoding the received subpacket (Step 216). If no error
is detected from the encoder packet acquired by the decoding in
step 216, i.e., if decoding of F-PDCH is successfully performed
("Yes" path from decision step 218), the mobile station transmits
ACK over R-ACKCH (Step 224). However, if the decoding performed in
step 216 failed ("No" path from decision step 218), the mobile
station increases the reception number Rx_Num by 1 (Step 220), and
transmits NAK over R-ACKCH (Step 226) while retaining the data
stored in the ARQ buffer (Step 222). Steps 212 to 222 in block 70
drawn in a dotted line represent a decoding process caused by
initial transmission (hereinafter referred to as "initial
transmission-related decoding process"), performed by the mobile
station.
However, if it is determined in step 210 that SP_ID is not set to
`0` ("No" path from the decision step 210), the mobile station
compares EP_SIZE acquired by decoding the F-PDCCH with previously
received EP_SIZE (Step 228). If the EP_SIZEs are not identical to
each other ("No" path from decision step 228), the mobile station
gates off the R-ACKCH (Step 206), and then returns to step 202,
determining that it is not possible to restore an encoder packet
since a retransmitted subpacket is received before an initially
transmitted subpacket is received.
If, as a result of the comparison in step 228, it is determined
that the EP_SIZEs are identical to each other ("Yes" path from
decision step 228), the mobile station code/packet-combines a
subpacket received over F-PDCH with previous subpackets previously
received and stored in the ARQ buffer, and stores the combined
packet in the ARQ buffer (Step 230). Thereafter, the mobile station
acquires an encoder packet by turbo-decoding the combined subpacket
(Step 232). If no error is detected from the encoder packet, i.e.,
decoding is successfully performed ("Yes" path from decision step
234), the mobile station transmits ACK over R-ACKCH (Step 244).
However, if the decoding is failed ("No" path from decision step
234), the mobile station increases the reception number Rx_Num by 1
(Step 236), and then compares the increased reception number Rx_Num
with a predetermined maximum reception number MAX_RX_NUM (Step
238). The maximum reception number MAX_RX_NUM prevents the mobile
station from endlessly waiting for retransmission of the same
encoder packet. If the reception number Rx_Num is larger than or
equal to the maximum reception number MAX_RX_NUM ("Yes" path from
decision step 238), the mobile station clears the ARQ buffer (Step
240), and transmits NAK over R-ACKCH (Step 246). If the reception
number Rx_Num is smaller than the maximum reception number
MAX_RX_NUM ("No" path from decision step 238), the mobile station
transmits NAK over R-ACKCH (Step 244) while retaining the ARQ
buffer (Step 242). The steps 230 to 242 in block 72 drawn by a
dotted line represent a decoding process caused by retransmission
(hereinafter referred to as "retransmission-related decoding
process"), performed by the mobile station.
FIGS. 9 and 10 are flowcharts illustrating H-ARQ operations by a
base station and a mobile station depending upon MAC_ID, ARQ_ID,
SP_ID, EP_SIZE and AI_SN according to an embodiment of the present
invention. Likewise, SP_ID=0 indicates initial transmission on a
first subpacket of a new encoder packet.
Referring to FIG. 9, a base station sets a transmission number
Tx_Num to 0 (Step 300), and determines whether the transmission
number Tx_Num is 0 in order to start a packet data service (Step
302).
If the transmission number Tx_Num is 0 ("Yes" path from decision
step 302), the base station toggles an AI_SN value, i.e., changes
the AI_SN value from 0 to 1 or from 1 to 0 (Step 304), transmits
control information including MAC_ID, ARQ_ID, SP_ID (herein SP_ID
is set to 0), EP_SIZE and the toggled AI_SN over F-PDCCH, while
transmitting an initial transmission subpacket of a new encoder
packet to a mobile station over F-PDCH (Step 306). The AI_SN,
together with the EP_SIZE, serves as a double checker that
partially compensates for any ambiguity in the sequence of encoder
packets. A mobile station determines whether current transmission
is retransmission on the same encoder packet, using both the AI_SN
and the EP_SIZE.
Thereafter, the base station waits for a response from the mobile
station for the transmitted subpacket (Step 308). The base station
decodes R-ACKCH from the mobile station (Step 310), and analyzes a
response from the mobile station for the transmitted subpacket
(Step 312). If a response from the mobile station is ACK ("Yes"
path from decision step 312), the base station returns to step 302
to transmit a new encoder packet. Otherwise, if the response from
the mobile station is NAK or Silence ("No" path from decision step
312), the base station increases the transmission number Tx_Num by
1 in order to retransmit the same encoder packet (Step 314), and
then returns to step 302.
If it is determined in step 302 that the transmission number Tx_Num
is not 0 ("No" path from decision step 302), the base station
determines whether the last response received from the mobile
station over R-ACKCH is Silence (Step 316). If the last response
from the mobile station is Silence ("No" path from decision step
316), i.e., if there is no previous response, the base station
transmits control information including previous SP_ID over F-PDCCH
while retransmitting a previously transmitted subpacket over F-PDCH
(Step 320). However, if the last response from the mobile station
is not Silence, i.e., if there is a previous response ("Yes" path
from decision step 316), the base station increases SP_ID of
control information transmitted over F-PDCCH by 1, while
transmitting a next subpacket of a previously transmitted encoder
packet, over F-PDCH (Step 318). After both steps 318 and 320, the
base station waits for a response from the mobile station for the
subpacket transmitted in step 318 or 320 (Step 322). The base
station decodes R-ACKCH from the mobile station (Step 324), and
analyzes a response from the mobile station for the subpacket
transmitted in step 318 or 320 (Step 326). If a response from the
mobile station is ACK ("Yes" path from decision step 326), the base
station resets the transmission number Tx_Num to `0` in order to
transmit a new encoder packet (Step 332), and then returns to step
302. However, if a response from the mobile station is NAK or
Silence ("No" path from decision step 326), the base station
increases the transmission number Tx_Num by 1 (Step 328), and then
proceeds to step 330.
In step 330, the transmission number Tx_Num is compared with a
previously set maximum transmission number MAX_TX_NUM. If the
transmission number Tx_Num is larger than or equal to the maximum
transmission number MAX_TX_NUM ("Yes" path from decision step 330),
the base station resets the transmission number Tx_Num to `0` in
order to transmit a new encoder packet (Step 332), and then returns
to step 302. However, if the transmission number Tx_Num is smaller
than the maximum transmission number MAX_TX_NUM ("No" path from
decision step 330), the base station returns to step 302 without
changing the transmission number Tx_Num in order to retransmit the
same encoder packet.
Referring to FIG. 10, a mobile station initializes an ARQ buffer
and sets a reception number Rx_Num to 0 in order to receive a
packet data service (Step 400). Thereafter, the mobile station
decodes F-PDCCH from a base station (Step 402), and determines
whether a CRC error is detected from control information acquired
by decoding the F-PDCCH thereby determining whether the mobile
station was successful in decoding (Step 404). The control
information acquired by decoding the F-PDCCH includes MAC_ID,
SP_ID, EP_SIZE and AI_SN. If the decoding failed ("No" path from
decision step 404), R-ACKCH is gated off (Step 406). If it is
determined in step 404 that decoding of F-PDCCH was successfully
performed ("Yes" path from decision step 404), the mobile station
then further determines whether MAC_ID acquired by the decoding is
identical to its own MAC_ID, thereby determining whether the
F-PDCCH had any information transmitted to the mobile station
itself (Step 408). If they are not identical to each other ("No"
path from decision step 408), R-ACKCH is gated off (Step 406).
Otherwise, if the MAC_IDs are identical to each other in step 408
("Yes" path from decision step 408), the mobile station determines
whether SP_ID acquired by the decoding is set to `0` (Step 410). If
the SP_ID is set to `0` ("Yes" path from decision step 410), the
mobile station determines whether the AI_SN and EP_SIZE acquired by
the decoding are identical to previously received AI_SN and EP_SIZE
(Step 412). If the AI_SN and EP_SIZE are both identical to the
previously received AI_SN and EP_SIZE ("Yes" path from decision
step 412), the mobile station transmits ACK over R-ACKCH without
decoding F-PDCH (Step 426), determining that although a base
station normally received a previously transmitted encoder packet,
the base station failed to normally receive ACK transmitted from
the mobile station.
However, if both the AI_SN and EP_SIZE are not identical to the
previously received AI_SN and EP_SIZE ("No" path from decision step
412), the mobile station sets the reception number Rx_Num to `0`
(Step 414), and then clears the ARQ buffer (Step 416). Thereafter,
the mobile station stores a subpacket received over F-PDCH, in the
ARQ buffer, and acquires an encoder packet by decoding the received
subpacket (Step 418). If no error is detected from the encoder
packet acquired by the decoding in step 418, i.e., if decoding of
F-PDCH is successfully performed ("Yes" path from decision step
420), the mobile station transmits ACK over R-ACKCH (Step 426).
However, if the decoding performed in step 418 is failed ("No" path
from decision step 420), the mobile station increases the reception
number Rx_Num by 1 (Step 422), and transmits NAK over R-ACKCH (Step
428) while retaining the data stored in the ARQ buffer (Step 424).
The steps 414 to 424 in block 74 drawn by a dotted line represent
an initial transmission-related decoding process performed by the
mobile station.
However, if it is determined in step 410 that SP_ID is not set to
`0` ("No" path from decision step 410), the mobile station compares
AI_SN and EP_SIZE acquired by decoding the F-PDCCH with previously
received AI_SN and EP_SIZE (Step 430). If both the AI_SN and
EP_SIZE are not identical to the previously received AI_SN and
EP_SIZE ("No" path from decision step 430), it means that a
retransmitted subpacket is received before an initially received
subpacket was received. Therefore, the mobile station will "gate
off" the R-ACKCH in order to retransmit the initially transmitted
subpacket (Step 406), and then returns to step 402.
As a result of the comparison in step 430, if the AI_SN and EP_SIZE
are both identical to the previously received AI_SN and EP_SIZE
("Yes" path from decision step 430), the mobile station
code/packet-combines a subpacket received over F-PDCH with previous
subpackets previously received and stored in the ARQ buffer, and
stores the combined packet in the ARQ buffer (Step 432).
Thereafter, the mobile station acquires an encoder packet by
turbo-decoding the combined subpacket (Step 434). If no error is
detected from the encoder packet, i.e., decoding is successfully
performed ("Yes" path from decision step 436), the mobile station
transmits ACK over R-ACKCH (Step 446).
However, if the decoding is failed ("No" path from decision step
436), the mobile station increases the reception number Rx_Num by 1
(Step 438), and then compares the increased reception number Rx_Num
with a predetermined maximum reception number MAX_RX_NUM (Step
440). If the reception number Rx_Num is larger than or equal to the
maximum reception number MAX_RX_NUM ("Yes" path from decision step
440), the mobile station clears the ARQ buffer (Step 442), and
transmits NAK over R-ACKCH (Step 448). If the reception number
Rx_Num is smaller than the maximum reception number MAX_RX_NUM
("No" path from decision step 440), the mobile station transmits
NAK over R-ACKCH (Step 446) while retaining the ARQ buffer (Step
444). The steps 432 to 444 in block 76 drawn by a dotted line
represent a retransmission-related decoding process performed by
the mobile station.
With reference to FIGS. 7 to 10, operations of the base and mobile
stations have been described on the assumption that an initial
transmission subpacket with SP_ID=0 must be necessarily received.
Therefore, when an initial transmission packet is missing, it is
not possible to restore an encoder packet. In order to make up for
the defects, FIGS. 11 and 12 illustrate operations of the base and
mobile station, wherein the mobile station can decode an encoder
packet even before an initially transmitted subpacket is
received.
FIGS. 11 and 12 are based on the fact that although a mobile
station misses a first subpacket and performs decoding on its
succeeding subpackets, there is a relatively high probability that
an encoder packet can be successfully decoded. In FIGS. 11 and 12,
it is determined that the penalty of turbo decoding due to missing
(loss) of an initially transmitted subpacket is negligible. In
addition, FIGS. 11 and 12 illustrate an operation in which a base
station unconditionally transmits a retransmission subpacket
without determining presence/absence of a previous response when it
decides upon retransmission. However, in an alternative embodiment
of the present invention, when there is no previous response, a
base station can retransmit a previous subpacket with a simple
modification.
It will be assumed herein that the H-ARQ operation is performed
using MAC_ID, ARQ_ID, SP_ID, EP_SIZE and AI_SN, and SP_ID=0
indicates initial transmission on a first subpacket of a new
encoder packet.
Referring to FIG. 11, a base station sets a transmission number
Tx_Num to 0 (Step 500), and determines whether the transmission
number Tx_Num is 0 in order to start a packet data service (Step
502).
If the transmission number Tx_Num is 0 ("Yes" path from decision
step 502), the base station toggles an AI_SN value, i.e., changes
the AI_SN value from 0 to 1 or from 1 to 0 (Step 504), transmits
control information including MAC_ID, ARQ_ID, SP_ID (herein SP_ID
is set to 0), EP_SIZE and the toggled AI_SN over F-PDCCH while
transmitting an initial transmission subpacket of a new encoder
packet to a mobile station over F-PDCH (Step 506). Thereafter, the
mobile station waits for a response from the mobile station for the
transmitted subpacket (Step 508). The base station decodes R-ACKCH
from the mobile station (Step 510), and analyzes a response from
the mobile station for the transmitted subpacket (Step 512). If a
response from the mobile station is ACK ("Yes" path from decision
step 512), the base station returns to step 502 to transmit a new
encoder packet. Otherwise, if a response from the mobile station is
NAK or Silence ("No" path from decision step 512), the base station
increases the transmission number Tx_Num by 1 (Step 514), and then
returns to step 502.
If it is determined in step 502 that the transmission number Tx_Num
is not 0 ("No" path from decision step 502), the base station
increases the SP_ID of control information transmitted over F-PDCCH
by 1, while transmitting a next subpacket of a previously
transmitted encoder packet, over F-PDCH (Step 516).
Thereafter, the base station waits for a response from the mobile
station for the subpacket transmitted in step 516 (Step 518). The
base station decodes R-ACKCH from the mobile station (Step 520),
and analyzes a response from the mobile station for the subpacket
transmitted in step 516 (Step 522). If a response from the mobile
station is ACK ("Yes" path from decision step 522), the base
station resets the transmission number Tx_Num to `0` in order to
transmit a new encoder packet (Step 528), and then returns to step
502. However, if a response from the mobile station is NAK or
Silence ("No" path from decision step 522), the base station
increases the transmission number Tx_Num by 1 (Step 524), and then
proceeds to step 526.
In step 526, the transmission number Tx_Num is compared with a
previously set maximum transmission number MAX_TX_NUM. If the
transmission number Tx_Num is larger than or equal to the maximum
transmission number MAX_TX_NUM ("Yes" path from decision step 526),
the base station resets the transmission number Tx_Num to `0` in
order to transmit a new encoder packet (Step 528), and then returns
to step 502. However, if the transmission number Tx_Num is smaller
than the maximum transmission number MAX_TX_NUM ("No" path from
decision step 526), the base station returns to step 502 without
changing the transmission number Tx_Num in order to retransmit the
same encoder packet.
Referring to FIG. 12, a mobile station initializes an ARQ buffer
and sets a reception number Rx_Num to 0 in order to receive a
packet data service (Step 600). Thereafter, the mobile station
decodes F-PDCCH from a base station (Step 602), and determines
whether a CRC error is detected from control information acquired
by decoding the F-PDCCH, thereby determining whether the mobile
station made a successful decoding (Step 604). The control
information acquired by decoding the F-PDCCH includes MAC_ID,
SP_ID, EP_SIZE and AI_SN. If the decoding failed ("No" path from
decision step 604), R-ACKCH is gated off (Step 606). If it is
determined in step 604 that decoding of F-PDCCH is successfully
performed ("Yes" path from decision step 604), the mobile station
determines whether MAC_ID acquired by the decoding is identical to
its own MAC_ID, thereby determining whether the F-PDCCH has any
information transmitted to the mobile station itself (Step 608). If
they are not identical to each other ("No" path to decision step
608), R-ACKCH is gated off (Step 606).
Otherwise, if the MAC_IDs are identical to each other in step 608
("Yes" path from decision step 608), the mobile station determines
whether SP_ID acquired by the decoding is set to `0` that indicates
initial transmission (Step 610). If the SP_ID is set to `0` ("Yes"
path from decision step 610), the mobile station determines whether
the AI_SN and EP_SIZE acquired by the decoding are identical to
previously received AI_SN and EP_SIZE (Step 612). If the AI_SN and
EP_SIZE are both identical to the previously received AI_SN and
EP_SIZE, ("Yes" path from decision step 612) the mobile station
transmits ACK over R-ACKCH (Step 626).
However, if both the AI_SN and EP_SIZE are not identical to the
previously received AI_SN and EP_SIZE ("No" path from decision step
612), the mobile station sets the reception number Rx_Num to `0`
(Step 614), and then clears the ARQ buffer (Step 616). Thereafter,
the mobile station stores a subpacket received over F-PDCH, in the
ARQ buffer, and acquires an encoder packet by decoding the received
subpacket (Step 618). If no error is detected from the encoder
packet acquired by the decoding in step 618, i.e., if decoding of
F-PDCH is successfully performed ("Yes" path from decision step
620), the mobile station transmits ACK over R-ACKCH (Step 626).
However, if the decoding performed in step 618 is failed ("No" path
from decision step 620), the mobile station increases the reception
number Rx_Num by 1 (Step 622), and transmits NAK over R-ACKCH (Step
628) while retaining the data stored in the ARQ buffer (Step 624).
The steps 614 to 624 in block 78 drawn by a dotted line represent
an initial transmission-related decoding process performed by the
mobile station.
However, if it is determined in step 610 that SP_ID is not set to
`0` ("No" path to decision step 610), the mobile station compares
AI_SN and EP_SIZE acquired by decoding the F-PDCCH with previously
received AI_SN and EP_SIZE (Step 630). If both the AI_SN and
EP_SIZE are not identical to the previously received AI_SN and
EP_SIZE ("No" path from decision step 630), it means that a
retransmitted subpacket is received before an initially received
subpacket was received. In this case, an encoder packet is decoded
using only the next subpacket without the initial subpacket.
Therefore, the mobile station proceeds to step 612 to perform
initial transmission instead of gating off the R-ACKCH.
As a result of the comparison in step 630, if the AI_SN and EP_SIZE
are both identical to the previously received AI_SN and EP_SIZE
("Yes" path from decision step 630), the mobile station
code/packet-combines a subpacket received over F-PDCH with previous
subpackets previously received and stored in the ARQ buffer, and
stores again the combined packet in the ARQ buffer (Step 632).
Thereafter, the mobile station acquires an encoder packet by
turbo-decoding the combined subpacket (Step 634). If no error is
detected from the encoder packet, i.e., decoding is successfully
performed ("Yes" path from decision step 636), the mobile station
transmits ACK over R-ACKCH (Step 646).
However, if the decoding failed ("No" path from decision step 636),
the mobile station increases the reception number Rx_Num by 1 (Step
638), and then compares the increased reception number Rx_Num with
a predetermined maximum reception number MAX_RX_NUM (Step 640). If
the reception number Rx_Num is larger than or equal to the maximum
reception number MAX_RX_NUM ("Yes" path from decision step 640),
the mobile station clears the ARQ buffer (Step 642), and transmits
NAK over R-ACKCH (Step 648). If the reception number Rx_Num is
smaller than the maximum reception number MAX_RX_NUM ("No" path
from decision step 640), the mobile station transmits NAK over
R-ACKCH (Step 648) while retaining the ARQ buffer (Step 644). The
steps 632 to 644 in block 80 drawn by a dotted line represent a
retransmission-related decoding process performed by the mobile
station.
Since transmission of forward data traffic over F-PDCH is
controlled through transmission of a retransmission-related
response (ACK or NAK) achieved between a base station and a mobile
station, the transmission of retransmission-related response must
be performed with high reliability. However, transmission through
F-PDCCH and R-ACKCH is achieved in a wireless channel environment,
causing a transmission error. Although an error occurs in F-PDCCH
or R-ACKCH, a mobile station should be able to handle packet data
traffic and transmit a suitable response according thereto.
From this point of view, with reference to FIGS. 13A to 13D and
FIGS. 14A to 14E, consideration will be taken regarding normal and
abnormal situations, which may occur during transmission and
reception of data packet and a response according to an H-ARQ
protocol. Herein, a transmission packet will be represented by
[A,0], where A means the contents of a corresponding encoder packet
and `0` means AI_SN of the encoder packet. In addition, first (or
initial) transmission subpackets will be represented by A,B,C;
second subpackets (or first retransmission subpackets) by A',B',C';
and third subpackets (or second retransmission subpackets) by
A'',B'',C''.
FIGS. 13A to 13D are signal flow diagrams illustrating normal
situations which may occur during transmission of data packet and a
response according to an H-ARQ protocol.
Referring to FIG. 13A, a base station (BS) transmits an initial
transmission subpacket [A,0] of a transmission encoder packet and
its associated control information over F-PDCH and F-PDCCH (Step
700). If a mobile station (MS) receives normally both the F-PDCH
and the F-PDCCH, i.e., if no error is detected from data and
information generated by decoding the F-PDCH and F-PDCCH, the
mobile station generates an ACK indicating normal receipt of the
encoder packet and transmits the ACK over R-ACKCH (Step 702). The
base station then transmits a next encoder packet [B,1] and its
associated control information over F-PDCH and F-PDCCH (Step 704).
If at least the F-DCCH is received normally, the mobile station
generates ACK or NAK according to the decoding result for the
F-PDCH and transmits the ACK or NAK over R-ACKCH (Step 706).
Referring to FIG. 13B, a base station (BS) transmits an initial
transmission subpacket [A,0] of a transmission encoder packet and
its associated control information over F-PDCH and F-PDCCH (Step
710). If a mobile station (MS) receives normally both the F-PDCH
and the F-PDCCH, the mobile station generates ACK and transmits the
ACK over R-ACKCH (Step 712). The base station then transmits a next
encoder packet [B,1] and its associated control information over
F-PDCH and F-PDCCH (Step 714). If both the F-PDCH and the F-PDCCH
are not received normally due to variation in channel condition or
movement of the mobile station, the mobile station can transmit no
response (Step 716). Even though there is no response, the base
station performs the operation as if a NAK was received. As a
result, the base station transmits a retransmission subpacket
[B',1] of the encoder packet B over F-PDCH, and at the same time,
transmits its associated control information over F-PDCCH (Step
718). SP_ID included in the control information for the
retransmission subpacket [B',1] is set to `1`. If the F-PDCCH
transmitted in step 718 is received normally, the mobile station
generates ACK or NAK according to the decoding result for F-PDCH
and transmits the ACK or NAK over R-ACKCH (Step 720).
FIG. 13C is a signal flow diagram illustrating a situation where a
mobile station waits for a retransmission packet after failure to
decode F-PDCH. Referring to FIG. 13C, a base station transmits an
initial transmission subpacket [A,0] of a transmission encoder
packet and its associated control information over F-PDCH and
F-PDCCH (Step 730). If F-PDCCH is received normally and an error is
detected from data acquired by decoding F-PDCH using the control
information, the mobile station generates an ACK message and
transmits it over R-ACKCH (Step 732). The base station then
transmits a retransmission subpacket [A',0] for the encoder packet
and its associated control information over F-PDCH and F-PDCCH
(Step 734). If at least the F-PDCCH is received normally, the
mobile station generates an ACK or NAK message according to the
decoding result for the F-PDCH, and transmits the ACK or NAK
message over R-ACKCH (Step 736).
FIG. 13D is a signal flow diagram illustrates a situation where
decoding of F-PDCH and F-PDCCH has failed. Referring to FIG. 13D, a
base station transmits an initial transmission subpacket [A,0] of a
transmission encoder packet and its associated control information
over F-PDCH and F-PDCCH (Step 740). If the F-PDCCH is received
normally and an error is detected from data acquired by decoding
the F-PDCH using the control information, the mobile station
generates a NAK message and transmits it over R-ACKCH (Step 742).
The base station then transmits a retransmission subpacket [A',0]
for the encoder packet and its associated control information over
F-PDCH and F-PDCCH (Step 744). SP_ID included in control
information for the retransmission subpacket [A',0] is set to `1`.
If both the F-PDCH and the F-PDCCH are not received normally due to
deterioration of channel conditions or movement of the mobile
station, the mobile station transmits no response (Step 746). Even
though there is no response, the base station performs the
operation as if the NAK was received. As a result, the base station
transmits a third subpacket (or second retransmission subpacket)
[A'',0] of the encoder packet A and its associated control
information over F-PDCH and F-PDCCH (Step 748). SP_ID included in
the control information for the third transmission subpacket
[A'',0] is set to `2`. If at least the F-PDCCH transmitted in step
748 is received normally, the mobile station generates an ACK or
NAK message according to the decoding result for F-PDCH, and
transmits the ACK or NAK over R-ACKCH (Step 720).
Next, abnormal situations will be described with reference to FIGS.
14A to 14E.
FIG. 14A is a signal flow diagram illustrating a situation where a
mobile station fails to decode F-PDCH, incorrectly transmits a
response thereto, and fails to receive an initial transmission
subpacket. Referring to FIG. 14A, a base station (BS) transmits an
initial transmission subpacket [A,0] of a transmission encoder
packet and its associated control information over F-PDCH and
F-PDCCH (Step 800). If the F-PDCCH is received normally and an
error exists in an encoder packet acquired by decoding the F-PDCH,
the mobile station (MS) transmits a NAK message over R-ACKCH (Step
802). At this point, although the mobile station transmitted the
NAK, the base station may mistake the NAK for an ACK due to
deterioration of the channel condition (Step 804). In this case,
although the mobile station actually transmitted NAK, the base
station determines that an ACK was received.
As a result, the base station transmits an initial transmission
subpacket [B,1] of a new encoder packet B and its associated
control information over F-PDCH and F-PDCCH (Step 806). At this
point, AI_SN is changed from `0` to `1` in order to indicate
transmission of a next encoder packet. The mobile station may fail
to receive both the new initial transmission subpacket [B',1] and
its associated control information and thus, may transmit no
response (Step 808). The base station then transmits a
retransmission subpacket [B',1] of the last transmitted subpacket
[B,1] as if the NAK was received (Step 810). In this case, the
mobile station handles separately the first encoder packet A and
the second encoder packet B, and transmits an ACK or NAK in
response thereto (Step 812). A detailed description of this will be
discussed below.
FIG. 14B is a signal flow diagram illustrating a situation where a
mobile station fails to decode F-PDCH, fails to receive a
retransmission subpacket, and incorrectly transmits a response
thereto. Referring to FIG. 14B, the base station (BS) transmits an
initial transmission subpacket [A,0] of a transmission encoder
packet and its associated control information over F-PDCH and
F-PDCCH (Step 820). If the F-PDCCH is received normally and an
error exists in an encoder packet received over the F-PDCH, the
mobile station (MS) transmits a NAK message over R-ACKCH (Step
822). In response to the NAK, the base station transmits a
retransmission subpacket [A',0] of the encoder packet A to the
mobile station (Step 824). If both the F-PDCH and the F-PDCCH
transmitted in step 824 are not received normally, the mobile
station transmits no response (Step 826). At this point, although
the mobile station transmitted no response, the base station may
detect an ACK due to deterioration of the channel condition (Step
828). As a result, the base station transmits an initial
transmission subpacket [B,1] of the next encoder packet instead of
transmitting a subpacket [A'',0] that should have been transmitted
(Step 830). The mobile station then handles separately the first
encoder packet A and the second encoder packet B, and transmits an
ACK or NAK in response thereto (Step 832). A detailed description
of this will be discussed below.
FIG. 14C is a signal flow diagram illustrating a situation where a
mobile station succeeds in decoding F-PDCH and F-PDCCH, and
incorrectly transmits an ACK message. Referring to FIG. 14C, a base
station (BS) transmits an initial transmission subpacket [A,0] of a
transmission encoder packet and its associated control information
over F-PDCH and F-PDCCH (Step 840). If both the F-DPCH and the
F-PDCCH are received normally, the mobile station (MS) generates an
ACK message and transmits it over R-ACKCH (Step 842). At this
point, although the mobile station transmitted the ACK, the base
station may detect a NAK due to deterioration of the channel
condition (Step 844). As a result, the base station transmits a
retransmission subpacket [A',0] of the encoder packet and its
associated control information over F-PDCH and F-PDCCH (Step 846).
If both the F-PDCH and the F-PDCCH transmitted in step 846 are not
received normally, the mobile station transmits no response (Step
848). If the mobile station transmits no response, the base station
performs the operation as if a NAK was received. As a result, the
base station transmits a third subpacket [A'',0] of the encoder
packet (Step 850). Even when retransmission is performed on the
previously successfully received encoder packet, the mobile station
separately generates an artificial ACK and transmits the artificial
ACK to the base station (Step 852). A detailed description of this
will be discussed below.
FIG. 14D is a signal flow diagram illustrating a situation where a
mobile station receives an initial transmission subpacket of a new
encoder packet before F-PDCH and F-PDCCH are received. Referring to
FIG. 14D, a base station (BS) transmits an initial transmission
subpacket [A,0] of a transmission encoder packet and its associated
control information over F-PDCH and F-PDCCH (Step 860). If both the
F-DPCH and the F-PDCCH are received normally, the mobile station
(MS) generates an ACK message and transmits it over R-ACKCH (Step
862). The base station then transmits an initial transmission
subpacket [B,1] of the next encoder packet and its associated
control information over F-PDCH and F-PDCCH (Step 864). If both the
F-PDCH and the F-PDCCH are not received normally, the mobile
station transmits no response (Step 866). Although the mobile
station transmitted no response, the base station may detect an ACK
message due to variations in the channel condition, movement of the
mobile station, or an interference signal from another mobile
station (Step 868). In this case, the base station transmits an
initial transmission subpacket [C,0] of the next encoder packet
(Step 870). The mobile station then handles separately the first to
third encoder packets A, B and C, and transmits ACK or NAK as a
response thereto (Step 872). A detailed description of this will be
discussed below.
FIG. 14E is a signal flow diagram illustrating a situation, where,
although a mobile station transmitted a NAK message indicating
failure to decode F-PDCH, a base station transmits a new encoder
packet by mistaking the NAK for an ACK indicating success in
decoding F-PDCH, and although the mobile station failed to receive
the new encoder packet, the base station transmits another new
encoder packet. Referring to FIG. 14E, the base station (BS)
transmits an initial transmission subpacket [A,0] of a transmission
encoder packet and its associated control information over F-PDCH
and F-PDCCH (Step 880). If the F-DPCCH is received normally and
there exists an error in an encoder packet received over the
F-PDCH, the mobile station (MS) transmits a NAK over R-ACKCH (Step
882). At this point, although the mobile station transmitted the
NAK, the base station may detect an ACK due to deterioration of the
channel condition (Step 884). As a result, the base station
transmits an initial transmission subpacket [B,1] of the next
encoder packet and its associated control information over F-PDCH
and F-PDCCH (Step 886). If both the F-PDCH and the F-PDCCH
transmitted in step 886 are not received normally, the mobile
station transmits no response (Step 888). However, although the
mobile station transmitted no response, the base station may detect
an ACK message due to variations in the channel condition or an
interference signal from another mobile station (Step 890). In this
case, the base station transmits an initial transmission subpacket
[C,0] of another next encoder packet (Step 892). The mobile station
then receives another new encoder packet C before it succeeds in
decoding the first encoder packet A and receives the second encoder
packet B. The mobile station handles separately the encoder packets
A, B and C, and transmits an ACK or NAK in response thereto (Step
894). A detailed description of this will be discussed below.
A description will now be made of the normal situations.
In the case where an operation is performed normally as described
in conjunction with FIGS. 13A to 13D, each time a subpacket and its
associated control information are received, the mobile station
compares a previously received AI_SN with a currently received
AI_SN, and determines whether the currently received subpacket is a
new subpacket (i.e., initial transmission subpacket) or a continue
subpacket (i.e., retransmission subpacket), as follows:
(1) if the current AI_SN is not identical to the previous AI_SN,
the mobile station determines that a received subpacket is an
initial transmission subpacket;
(2) if the current AI_SN is identical to the previous AI_SN, the
mobile station determines that a received subpacket is a
retransmission subpacket.
In the case where no error exists in R-ACKCH, when the mobile
station transmits ACK over R-ACKCH and then expects to receive an
initial transmission subpacket, the base station actually transmits
an initial transmission subpacket. Likewise, when the mobile
station transmits a NAK over R-ACKCH and then expects to receive a
retransmission subpacket, the base station actually transmits a
retransmission subpacket. Therefore, upon receiving a
retransmission subpacket, the mobile station performs a
retransmission-related decoding process on the received subpacket.
The retransmission-related decoding process, as mentioned above,
includes code/packet-combining and decoding.
In the case where an error has occurred in R-ACKCH, as described in
conjunction with FIGS. 14A to 14E, the mobile station compares a
currently received AI_SN with a previously received AI_SN, and
determines whether a currently received subpacket is a new
subpacket (i.e., initial transmission subpacket) or a continue
subpacket (i.e., retransmission subpacket), as follows:
(1) If the current AI_SN is not identical to the previous AI_SN,
the mobile station determines that a received subpacket is an
initial transmission subpacket;
(2) If the current AI_SN is identical to the previous AI_SN, the
mobile station determines whether a received subpacket is an
initial transmission subpacket or a retransmission subpacket by
applying additional criteria for more accurate determination.
In abnormal situations, particularly, when the current AI_SN is
identical to a previous AI_SN, the mobile station cannot determine
whether a received subpacket is a retransmission subpacket by
simply comparing the AI_SNs for the following reasons.
If the base station incorrectly analyzes or fails to receive the
ACK/NAK transmitted over R-ACKCH by a mobile station due to an
error occurring in the R-ACKCH, the base station transmits another
subpacket rather than a subpacket that the mobile station expects
to receive. In this case, when the mobile station expects an
initial transmission subpacket, the base station transmits a
retransmission subpacket. Alternatively, when the mobile station
expects a retransmission subpacket, the base station transmits an
initial transmission subpacket. This corresponds to the situations
of FIGS. 14A and 14B, wherein the ACK and NAK message were mistaken
for NAK and ACK messages respectively.
In the situations of FIGS. 14A and 14B, a currently received
subpacket is regarded as an initial transmission subpacket by the
mobile station, and undergoes an initial transmission-related
decoding process. The initial transmission-related decoding
process, as mentioned above, includes resetting of an ARQ buffer
and storing and decoding of a currently received subpacket.
Even in the case where an error has occurred in R-ACKCH, if the
received AI_SN is not identical to the previous AI_SN, the mobile
station regards a currently received subpacket as an initial
transmission subpacket. This is because it is possible to restore
an encoder packet with only a retransmission subpacket without an
initial transmission subpacket.
With reference to FIGS. 14C to 14E, a description will now be
provided regarding the situations wherein the current AI_SN is
identical to a previous AI_SN.
In the situation illustrated in FIG. 14C, a currently received
subpacket was generated from an encoder packet A, and an ACK
message was transmitted after decoding the encoder packet A
successfully. Therefore, the mobile station transmits an artificial
ACK over R-ACKCH without decoding the currently received
subpacket.
In the situation illustrated in FIG. 14D, although an ACK/NAK
message was not transmitted over R-ACKCH, the base station detects
an ACK message. A subpacket currently received by the mobile
station was generated from a new encoder packet C. In this case,
the mobile station regards the currently received subpacket as an
initial transmission subpacket in performing an initial
transmission-related decoding process. In this case, the subpacket
[B,1] of a second encoder packet B is disregarded, since
retransmission request for the subpacket [B,1] will be performed by
an upper layer.
Finally, in the situation illustrated in FIG. 14E, when a mobile
station waits for a retransmission subpacket after transmitting a
NAK message over R-ACKCH, a subpacket is received whose current
AI_SN and previous AI_SN are identical to each other. This
corresponds to the case in which after the decoding of F-PDCCH
failed NAK was mistakenly transmitted over R-ACKCH as ACK. The
currently received subpacket was generated from a new encoder
packet C. Therefore, the mobile station determines that the
currently received subpacket is an initial transmission subpacket,
and performs an initial transmission-related decoding process.
Likewise, a subpacket [A,0] of a first subpacket A and a subpacket
[B,1] of a second encoder packet B are disregarded, since
retransmission request for the subpackets [A,0] and [B,1] will be
performed by an upper layer.
In summary, if AI_SN is not changed in a state where an error has
occurred in R-ACKCH, the mobile station operates as follows. In
particular, when the mobile station was waiting for an initial
transmission subpacket, the mobile station must conduct different
operations according to the type of error which occurred in the
R-ACKCH. The following three (3) examples illustrate the
aforementioned types of errors.
(1) If a current AI_SN is identical to a previous AI_SN while the
mobile station waits for an initial transmission subpacket, the
mobile station transmits an artificial ACK over R-ACKCH since the
received subpacket is a retransmission subpacket.
(2) If a current AI_SN is not identical to a previous AI_SN while
the mobile station waits for an initial transmission subpacket, the
mobile station performs an initial transmission-related decoding
process since the received subpacket is an initial transmission
subpacket.
(3) If a current AI_SN is identical to a previous AI_SN while the
mobile station waits for a retransmission subpacket, the mobile
station performs an initial transmission-related decoding process
since the received subpacket is an initial transmission
subpacket.
A mobile station determines whether to retransmit a subpacket by
applying an H-ARQ protocol in the following two methods: First, if
the current AI_SN is not identical to a previous AI_SN, the mobile
station determines that the received subpacket is an initial
transmission subpacket regardless of whether an error has occurred
in R-ACKCH. Second, in the case where the current AI_SN is
identical to a previous AI_SN, the mobile station determines that
the received subpacket is a retransmission subpacket if no error
exists in R-ACKCH. However, if an error exists in R-ACKCH, the
mobile station determines that the received subpacket is an initial
transmission subpacket or a retransmission subpacket, according to
the type of the error.
However, the mobile station cannot correctly determine whether an
error has occurred in R-ACKCH, and the type of the error, if it
occurred. Therefore, the mobile station minimizes
retransmission-related mis-operation by predicting whether an error
has occurred in R-ACKCH, depending upon SP_ID and EP_SIZE, as
follows:
(A) If the current AI_SN is not identical to the previous AI_SN,
the mobile station determines that the received subpacket is an
initial transmission subpacket;
(B) If the current AI_SN is identical to the previous AI_SN, the
mobile station compares the currently received EP_SIZE with the
previously received EP_SIZE;
(B-1) If the current EP_SIZE is not identical to the previous
EP_SIZE, the mobile station determines that the received subpacket
is an initial transmission subpacket; and
(B-2) If the current EP_SIZE is identical to the previous EP_SIZE,
the mobile station determines whether a response previously
transmitted over R-ACKCH is an ACK or NAK. If an ACK was previously
transmitted, it means that the mobile station was waiting for an
initial transmission subpacket. If a NAK was previously
transmitted, it means that the mobile station was waiting for a
retransmission subpacket. If the current AI_SN is identical to the
previous AI_SN while the mobile station is waiting an initial
transmission subpacket, it implies an abnormal situation in which
an error has occurred in R-ACKCH. If the current AI_SN is identical
to the previous AI_SN while the mobile station is waiting for a
retransmission subpacket, it implies a normal situation in which no
error has occurred in R-ACKCH.
Meanwhile, in the abnormal situation in which the current AI_SN is
identical to the previous AI_SN when a mobile station is waiting
for a retransmission subpacket, the mobile station should determine
that a received subpacket is an initial transmission subpacket and
perform the corresponding process. Actually, however, the mobile
station determines that the currently received subpacket is a
retransmission subpacket, considering the current situation as a
normal situation in which no error has occurred in R-ACKCH.
However, in FIG. 4E, if the size EP_SIZE of an encoder packet C is
larger than the size (previous EP_SIZE) of an encoder packet A, the
mobile station determines that the currently received subpacket is
an initial transmission subpacket previously received by (B-1). In
addition, since there is very low probability that the event of
FIG. 4E will happen, there is almost no initial transmission
subpacket that cannot be detected even through comparison between
the current EP_SIZE and the previous EP_SIZE. Therefore, in this
case, the following two determinations are available:
(a) If a previous response is a NAK, the mobile station determines
that the received subpacket is a retransmission subpacket.
(b) If a previous response is an ACK, the mobile station determines
whether the received SP_ID is set to 0 or not.
Although the mobile station recognizes an abnormal situation when
the current AI_SN is identical to the previous AI_SN and a previous
response is ACK, the mobile station cannot determine what type of
an error has occurred in R-ACKCH. Therefore, the mobile station
cannot determine whether a received subpacket is a retransmission
subpacket, for which an artificial ACK is required, or an initial
transmission subpacket. An initial transmission subpacket, which
cannot be determined even through comparison of EP_SIZE, can be
detected by SP_ID. However, in some cases, there exists a
retransmission subpacket with SP_ID=0. Therefore, if subpackets
with SP_ID=0 are all determined as initial transmission subpackets,
then a retransmission subpacket, for which an ACK is required, may
be occasionally incorrectly determined as an initial transmission
subpacket.
If SP_ID.noteq.0, it is preferable that the mobile station
transmits an artificial ACK over R-ACKCH without decoding the
received subpacket. However, as mentioned above, the mobile station
can succeed in decoding an encoder packet through a
retransmission-related decoding process with only a retransmission
subpacket without a general initial transmission subpacket.
Therefore, the present invention does not separately distinguish a
retransmission subpacket for which artificial ACK is required, from
a general subpacket.
FIG. 15 is a flowchart illustrating an H-ARQ operation by a mobile
station according to an embodiment of the present invention. The
procedure described below is performed after the decoding of
F-PDCCH and comparison of MAC_ID have been accomplished, described
in conjunction with FIGS. 8, 10 and 12.
Referring to FIG. 15, a mobile station compares a currently
received AI_SN with a previously stored AI_SN, thereby determining
whether an encoder packet is received whose AI_SN is identical to
the previous AI_SN (Step 910). If an encoder packet is received
which has an AI_SN which is not identical to the previous AI_SN
("No" path from decision step 910), the mobile station performs an
initial transmission-related decoding process (Step 970). In step
970 the mobile station clears an ARQ buffer, acquires an encoder
packet by decoding an initial subpacket received over F-PDCH, and
determines whether an error is detected from the encoder packet. If
no error is detected, the mobile station transmits an ACK. If an
error is detected, the mobile station transmits a NAK. When a NAK
is transmitted, the initial subpacket is stored in the ARQ buffer.
As a result of the determination in step 910, if the current AI_SN
is identical to the previous AI_SN ("Yes" path from decision step
910), the mobile station determines whether a currently received
EP_SIZE is identical to a previously received EP_SIZE (Step 920).
If the two EP_SIZEs are not identical to each other ("No" path from
decision step 920), the mobile station performs the initial
transmission-related decoding process, determining that a new
encoder packet was received (Step 970).
However, if the two EP_SIZEs are identical to each other ("Yes"
path from decision step 920), the mobile station determines whether
a response previously transmitted over R-ACKCH is an ACK (Step
930). If a previously transmitted response is NAK ("No" path from
decision step 930), the mobile station performs a
retransmission-related decoding process (Step 950). In step 950 the
mobile station combines a subpacket received over F-PDCH with
subpackets previously stored in the ARQ buffer, acquires an encoder
packet by decoding the combined subpacket, and determines whether
an error is detected from the encoder packet. If no error is
detected, the mobile station transmits an ACK. However, if an error
is detected, the mobile station transmits a NAK. When a NAK is
transmitted, the combined subpacket is stored in the ARQ buffer.
However, if a reception number for the same encoder packet has
arrived at a previously determined maximum reception number, the
mobile station abandons reception of the encoder packet and clears
the ARQ buffer.
If it is determined in step 930 that a previously transmitted
response is an ACK ("Yes" path from decision step 930), the mobile
station determines whether a currently received SP_ID is set to 0
(Step 940). If the SP_ID is set to 0 ("Yes" path from decision step
940), the mobile station performs the initial transmission-related
decoding process (Step 970). Otherwise, the mobile station performs
the retransmission-related decoding process ("No" path from
decision step 950).
After performing the initial transmission-related decoding process
(Step 970) and the retransmission-related decoding process (Step
950), the mobile station stores currently received control
information AI_SN, EP_SIZE and ACK/NAK response (Step 960).
The present invention has been described to include at least the
following advantages:
In a 1.times.EV-DV high-speed wireless packet data communication
system, the present invention efficiently performs H-ARQ,
contributing to prevention of ambiguity which may occur during
system realization, and improvement of system throughput. In
addition, the present invention promptly properly handles received
subpackets regardless of whether an error has occurred in R-ACKCH,
enabling a packet data service at higher speed.
While the invention has been shown and described with reference to
a certain preferred embodiment thereof, it will be understood by
those skilled in the art that various changes in form and details
may be made therein without departing from the spirit and scope of
the invention as defined by the appended claims.
* * * * *